METALLURGICAL AND MATERIALS TRANSACTIONS B
ABSTRACTS
Volume 28B, No. 2, April 1997

This Month Featuring: Hydrometallurgy, Pyrometallurgy, Transport Phenomena, Process Control, Physical Chemistry, Solidification, Solid State Reactions, and Mathematical Modeling. View April 1997 Table of Contents.

HYDROMETALLURGY

Synergistic Extraction of Iron(III) at Higher Concentrations in D2EHPA-TBP Mixed Solvent Systems
K.K. SAHU and R.P. DAS
The extraction of iron(III) from chloride solutions at macrolevel concentration by different solvents such as tri-n-butyl phosphate (TBP), di(2-ethylhexyl) phosphoric acid (D2EHPA), and their mixture in various proportions has been investigated at different acid concentrations. The synergistic extraction of iron(III) with a mixture of TBP and D2EHPA was studied and the results were compared with that of the extraction by individual solvent alone. An increase in the concentration of the synergist, TBP, in the D2EHPA-TBP solvent system resulted in an increase in the synergistic coefficient value. The experimental data are treated graphically to explain the formation of organic phase extracted species, and the equilibrium extraction constants for the species are determined. It is found that a maximum of two molecules of TBP are adducted to the extracted species of the corresponding nonsynergistic system. Stripping of iron(III) with hydrochloric acid from loaded D2EHPA was found to increase with an increase in acid concentration. In the case of D2EHPA- TBP mixtures, stripping efficiency was increased with an increase in acid concentration up to a certain level and then it was decreased. The experimental results indicate that an iron exchange reaction between loaded D2EHPA and TBP proceeds during stripping at a higher concentration of hydrochloric acid (from mixed loaded solvent system). A plausible mechanism for iron(III) extraction and stripping has been discussed.

Communication: Discussion of "The Adsorption kinetics of Dycyanoaurate and Dycyanoargentate ions in activated Carbon"
N.M. VEGTER and R.F. SANDENBERGH

PYROMETALLURGY

Copper Losses and Thermodynamic Considerations in Copper Smelting
R. SRIDHAR, J.M. TOGURI, and S. SIMEONOV
A relationship between copper in slag and copper in matte during copper sulfide smelting has been derived using industrial data from 42 plants employing blast furnaces, reverberatory furnaces, flash furnaces, and Mitsubishi smelting furnaces together with the available thermodynamic equilibrium data for Cu-Fe-S-O, FeO-SiO2, and Cu-Fe-S systems and laboratory slag-matte equilibrium information. A copper smelting diagram showing oxygen potential; sulfur potential; and copper, magnetite, and sulfur contents in slag during the smelting of different grades of copper mattes is developed for mattes containing less than 70 pct copper. The data presented can be used to determine the entrained copper losses in slag. Further, by combining the calculated value of the entrained matte with the corresponding plant data for the sulfur content of the slag, it is possible to derive the dissolved sulfur content of the slag. These calculated values were in excellent agreement with the experimentally determined sulfide capacity of fayalite slags. It is shown that there is no need to assume the presence of dissolved copper sulfide species in industrial slags. The existing equilibrium data that relate the copper content of slags to oxygen potential adequately describe the copper losses in industrial slags.

Rate Control of the Flash Reduction of Zinc Calcines
E.M. WEENINK and N.J. THEMELIS
This work was part of a feasibility study on the production of zinc by flash-reducing zinc calcines (approximately consisting of 60 pct Zn and 10 pct Fe). The rate of reduction of single pellets was determined and incorporated in a unidimensional mathematical model of the transport and chemical reaction phenomena in a flash reaction shaft. The final experimental stage was to flash-reduce zinc calcines in a 0.13-m i.d. x 1.2-m-high reaction shaft at feed rates of 1.5 to 3 kg/h and in a range of CO/CO2 mixtures, which represented the products of the carbon-oxygen reactions in an industrial reactor. It was found that the principal variable affecting the rate of reduction was the CO/CO2 ratio. The observed low rates of reduction of zinc calcine particles were in fair agreement with the projections of the mathematical model and were also confirmed by tests on a mini-pilot flash reactor (40 kg calcine/h), carried out by Outokumpu Research (Pori, Finland). The main conclusion was that the reduction process is chemically controlled at rates which are an order of magnitude less than the flash oxidation of sulfide concentrates. Therefore, the reaction shaft of a zinc flash reduction furnace must be considerably longer than for the conventional flash smelting of sulfide concentrates.

TRANSPORT PHENOMENA

Multiphase Oxidation of Metals
THEODORE A. STEINBERG, SUBHASISH SIRCAR, D. BRUCE WILSON, and JOEL M. STOLTZFUS
The burning of metals in enriched oxygen atmospheres includes multiple phases and, therefore, satisfies the definition of being a heterogeneous system. For a reaction to be considered heterogeneous, however, the site of the chemical transformation must be at an interface. While the Wagner theory of metal oxidation gives satisfactory mechanisms for this type of reaction for solid metals, no comparable theory is available for metal oxidation when the temperature is above the melting point of the metal. Similitude theory is applied to metal combustion under conditions of the NASA/ASTM flammability test system in order to identify rate-controlling regimes and conditions of heterogeneous reaction. Above 4 MPa, the observed burn rate for iron is proportional to the sample dimension, L, as L-0.7. Analysis shows that the heat transfer rate between the reaction surface and the solid rod is proportional to L-0.5 and may, therefore, be the rate-determining process for the system.

PROCESS CONTROL

Intelligent Control of the Feeding of Aluminum Electrolytic Cells Using Neural Networks
A. MEGHLAOUI, R.T. BUI, J. THIBAULT, L. TIKASZ, and R. SANTERRE
To be efficient, the control of alumina feeding of the electrolytic cell must be based on cell resistance, alumina concentration, and cell state. Most control schemes now in use are based on cell resistance only, and, thus, constitute an open-loop control that lacks robustness because their decision criteria are not explicitly tied to concentration nor to cell state. This results in the cell operating at nonoptimal concentrations, and cell efficiency is diminished. An optimal operation requires a knowledge of concentration and an adjustment of the decision criteria as a function of concentration. A learning vector quantization (LVQ) type of neural network was built and trained to recognize the cell state. Knowing the state of the cell and its resistance, concentration can be estimated using predetermined regression functions. The decision criteria for the control logic are then consequently adapted. A closed-loop control scheme is thus obtained. Results show that, with its control so structured, the cell can operate at or near optimal concentrations independently of its state. This flexible and intelligent character of the neural control can provide a considerable advantage as compared to the standard control.

PHYSICAL CHEMISTRY

Kinetics of Chlorination and Carbochlorination of Pure Tantalum and Niobium Pentoxides
E. ALLAIN, M. DJONA, and I. GABALLAH
Kinetics of chlorination and carbochlorination of pure Nb2O5 and Ta2O5 were studied by thermogravimetric analysis between 385°C and 1000°C using Cl2-N2 and Cl2-CO-N2 gas mixtures. Standard free energy changes of the reactions and phase stability diagrams of Nb-O-Cl and Ta-O-Cl systems were calculated. The chlorination reaction order, for both oxides, with respect to Cl2 in the Cl2-N2 gas mixture was 0.82. The apparent activation energies (Ea) for Nb2O5 chlorination were 208 and 86 kJ/mole for temperatures lower and higher than 850°C, respectively. The experimental data could be described by a shrinking sphere model between 700°C and 1000°C. The chlorination mechanism, between 700°C and 850°C, was likely controlled by the chemical reaction. For T > 850°C, the overall Nb2O5 chlorination rate was affected by the allotropic transformation of the Nb2O5 T form to M form. Between 925°C and 1000°C, Ea for Ta2O5 chlorination was 246 kJ/mole. In this case, the most appropriate model was also that of shrinking sphere suggesting that the chlorination of Ta2O5 was controlled by the chemical reaction. For both oxides, the carbochlorination reaction order with respect to Cl2 + CO partial pressure, in the gas mixture, was about 2. The mathematical analysis of carbochlorination data indicates that Nb2O5 and Ta2O5 reactions could be described by shrinking sphere or cylinder, respectively. Below 600°C, the Ea values of Nb2O5 and Ta2O5 carbochlorination were 74 and 110 kJ/mole, respectively. Chemical reaction was probably the rate controlling step in both cases. An anomaly characterized by a decrease of the reaction rates occurs in the Arrhenius plots between 600°C and 800°C. This anomaly could be attributed to the thermal decomposition of COCl2 formed in situ during the carbochlorination.

Critical Evaluation and Optimization of the Thermodynamic Properties and Phase Diagrams of the CrO-Cr2O3-SiO2-CaO System
SERGEI DEGTEROV and ARTHUR D. PELTON
Available thermodynamic and phase diagram data have been critically assessed for all phases in the CrO-Cr2O3-SiO2-CaO system from 298 K to above the liquidus temperatures at all compositions under reducing conditions and at low CaO concentrations under oxidizing conditions. All reliable data have been simultaneously optimized to obtain one set of model equations for the Gibbs energy of the liquid slag and all solid phases as functions of composition and temperature. The modified quasichemical model was used for the slag. The models permit phase equilibria to be calculated for regions of composition, temperature, and oxygen potential where data are not available.

Thermodynamics of Iron Oxide in FexO-Dilute CaO + Al2O3 + FexO Fluxes at 1873 K
HIROYASU FUJIWARA, MASAHIRO KITOU, TAKAYUKI MATSUI, and EIJI ICHISE
The distribution of iron between FexO-dilute CaO + Al2O3 + FexO fluxes and Pt + Fe alloys, as well as the redox equilibrium of iron ions in these fluxes, was experimentally investigated in pressure- controlled CO2/CO gas at 1873 K. Total iron content in flux (pct FeT) and the ratio of (pct Fe2+) to (pct FeT) in fluxes with constant XCaO/XAl2O3 can be related to the activity of iron, aFe, and the partial pressure of oxygen, pO2, using the following equation:

(pct FeT) = {(CFe2+)pO21/2 + (CFe3+) pO23/4)} aFe


where CFe2+ and CFe3+ are the ferrous and ferric ion capacities, respectively, defined as
The present article applies these parameters to the evaluation of the activity coefficient of FexO at infinite dilution, °FexO, relative to the liquid iron oxide in equilibrium with iron. Furthermore, the composition dependence of °FexO is discussed.

Galvanic Cell Measurements on Supersaturated Activities of Oxygen in Fe-Al-M (M = C, Te, Mn, Cr, Si, Ti, Zr, and Ce) Melts
GUANGQIANG LI and HIDEAKI SUITO
Using a mullite (3Al2O3 · 2SiO2)-tube and ZrO2-9 mol pct MgO-plug type solid electrolyte galvanic cells, the activities of supersaturated oxygen in Fe-0.0017 to 0.41 mass pct Al-M (M = C, Te, Mn, Cr, Si, Ti, Zr, and Ce) alloys were measured as a function of total Al or M contents at 1873 K in an alumina crucible. Based on these results, the effects of alloying elements on the supersaturated oxygen activity with respect to alumina precipitation were studied. In the Fe-Al-M (M = C, Te, Mn, Cr, and Si) alloys, the supersaturated oxygen activities for a given Al level approach the equilibrium values with increasing contents of alloying elements. However, the oxygen activities for a given Al level are still supersaturated in the Fe-Al-M (M = Ti, Zr, and Ce) alloys even in the presence of considerable amounts of the alloying elements.

Effect of Alloying Element M (M = C, Te, Mn, Cr, Si, Ti, Zr, and Ce) on Supersaturation during Aluminum Deoxidation of Fe-Al-M Melts
GUANGQIANG LI and HIDEAKI SUITO
The effect of the alloying element M (M = C, Te, Mn, Cr, Si, Ti, Zr, And Ce) on the supersaturation of alumina precipitation in the Fe-0.0017 to 0.41 mass pct Al-M alloys was studied at 1873 K in an Al2O3 crucible based on the contents of analyzed aluminum and oxygen. It was found that the supersaturation ratio with respect ot the Al2O3 precipitation, S°Al2O3{=(aAl2 · aO3)obs/(aAl2 · aO3)eq}, decreased with increasing content of the alloying element in the Fe-Al-M (M = C, Te, Mn, Cr, and Si) alloys and approached unity in the range of [mass pct M] 0.2 to 0.5. In the Fe-0.011 to 0.41 mass pct Al-M (M = Ti, Zr, and Ce) alloys, however, the S°Al2O3 values were found to be independent of the contents of the alloying elements up to the compositions of [mass pct Ti] = 1.03, [mass pct Zr] = 0.08, and [mass pct Ce] = 0.07.

Reduction of Molybdenite with Carbon in the Presence of Lime
R. PADILLA, M.C. RUIZ, and H.Y. SOHN
The thermodynamics of the MoS2-C-CaO system has been studied in order to understand the carbothermic reduction of molybdenite in the presence of CaO. Kinetic studies were also conducted with mixtures of MoS2 + C + CaO in the temperature range of 900°C 1200°C. The reduction of MoS2 with carbon in the presence of lime proceeds through the direct oxidation of MoS2 by CaO to form intermediate molybdenum oxidized species, MoO2 and CaMoO4, which subsequently undergo reduction by CO to yield mixtures of Mo, Mo2C, and CaS. Complete conversion of MoS2 can be obtained at 1200°C in less than 20 minutes for molar concentrations of MoS2:C:CaO = 1:2:2. The kinetic model ln (1 - X) = kt was used to determine the rate constants. The activation energy found for the temperature range studied was 218.8 kJ/mol.

Dissolution of Alumina in Mold Fluxes
X. YU, R.J. POMFRET, and K.S. COLEY
The solubility and rate of dissolution of alumina in a range of mold fluxes in the CaO-Al2O3-SiO2-Na2O-CaF2 system have been measured at 1530°C using the rotating finger method. The solubilities were approximately 38 pct for all fluxes studied. The kinetics of dissolution were correlated with the Levich-Chochran equation for a rotating disc and used to determine the effective diffusivity of alumina. The effective diffusivity was inversely proportional to the viscosity of the flux and showed excellent agreement with previous work on CaO-Al2O3-SiO2 slags. This is in keeping with either the Eyring relation or the Stokes-Einstein relation. The value obtained for the size of the diffusing unit when the Stokes-Einstein relation was used, 1.81 Å, was more reasonable, which is not normally expected for silicate slags. This result is explained by the low level of polymerization in these fluxes.

SOLIDIFICATION

Calculation of Solidification-Related Thermophysical Properties for Steels
JYRKI MIETTINEN
Special algorithms have been developed to calculate important solidification-related thermophysical properties: enthalpy and enthalpy-related data (i.e., specific and latent heat), density, and thermal conductivity for low-alloyed and stainless steels. The algorithms are heavily based on the use of earlier developed phase transformation models, an interdendritic solidification model (IDS), and an austenite decomposition model (ADC), which solve, as a function of temperature, the phase fractions and compositions needed in these calculations. As a result, the thermophysical properties can be calculated at any temperature, from 1600°C to 25°C, taking into account the discontinuities caused by special phase transformations (i.e., ferritic, austenitic and peritectic solidification, ferrite/austenite transformation, and austenite decomposition to various structures) influenced by the steel grade and the cooling conditions.

SOLID STATE REACTIONS

Indentation Creep in Nanocrystalline Fe-TiN and Ni-TiN Alloys Prepared by Mechanical Alloying
YOSHIKIYO OGINO, TOHRU YAMASAKI, and BAO LONG SHEN
Mechanical properties of nanocrystalline Fe-TiN and Ni-TiN alloys with various TiN contents be tween 17 and 64 vol pct, which are prepared by dynamically consolidating mechanically alloyed powders, have been investigated by means of hardness measurements and indentation creep tests at intermediate temperatures. The hardness increases with decreasing grain size to about 10 nm. The indentation creep curves conform well to an equation derived from a transient creep rate equation. The analysis of creep curves indicates that the deformation occurs by a dislocation mechanism controlled by grain boundary diffusion.

Kinetics of Reduction of MnO in Powder Mixtures with Carbon
W.J. RANKIN and J.R. WYNNYCKYJ
Experimental data on the kinetics of the reaction between MnO and graphite in a mixture of fine powders, obtained earlier by one of the authors, were reassessed. The principal motive was to test on this system a new kinetic model, the "intrinsic transport" model, published previously by the present authors. In this model, the reaction-rate-limiting step is assumed to be transport of a gaseous intermediate by pore diffusion between reaction sites on the surfaces of the reactant particles. Various other model formalisms potentially applicable to this reaction also were tested. It is shown that the intrinsic transport of CO2 gave good to excellent fits to the experimental data over a wide range of reaction conditions. Other models gave poor agreement. It is shown, furthermore, that for the case where the CO2 pore-diffusion path lengths were made small (fine MnO and coarse graphite particle size), thus speeding up rapid intrinsic transport, the graphite-surface reaction became rate controlling.

MATHEMATICAL MODELING

Evaluation of Six k- Turbulence Model Predictions of Flowin a Continuous Casting Billet-Mold Water Model Using Laser Doppler Velocimetry Measurements
X.K. LAN, J.M. KHODADADI, and F. SHEN
Quantitative fluid velocity and turbulence measurements, which were obtained using laser Doppler velocimetry (LDV) during the course of the physical modeling of the molten steel flow in a contin uous casting billet mold, are presented. The experimental findings are compared to the results ob tained by the computational simulation of the flow using the high-Reynolds-number and five low-Reynolds-number k- turbulence models. The predictive capabilities of these turbulence models were assessed in comparison to the experimentally measured axial velocity and turbulence kinetic energy values. The measured turbulence velocity fluctuations in various directions strongly suggest the non isotropic nature of the flow field, which cannot be predicted with the k- models. The extent of deviations between the measurements and computations was quantified.

Modeling the Discontinuous Liquid Flow in a Blast Furnace
G.X. WANG, S.J. CHEW, A.B. YU, and P. ZULLI
This article presents a mathematical model to describe the discontinuous flow of an isothermal liquid in packed beds, simulating in part the flow condition in and below the blast furnace cohesive zone. The model is developed based on a force balance approach to describe the discrete liquid flow and a stochastic treatment to take into account the complex packing structure. The interaction between gas and liquid flows has also been included in the governing equations, so that the localized liquid flow in a packed bed can be modeled with or without gas flow. The difference between the micro scopic and macroscopic approaches is discussed, and it is argued that at this stage of development, liquid flow modeling should be conducted at the macroscopic level. Techniques for numerical so lution are provided. The validity of the proposed model is demonstrated by comparing model pre dictions with measurements obtained using a two-dimensional cold model apparatus under different gas and/or liquid flow conditions.

Direct questions about this or any other Metallurgical and Materials Transactions page to mettrans@andrew.cmu.edu.

Search TMS Document Center Tables of Contents Subscriptions Met. and Mat. Trans. TMS OnLine